Building Safety Inspections: Paper vs Electronic Inspections

Last registered on August 02, 2019

Pre-Trial

Trial Information

General Information

Title
Building Safety Inspections: Paper vs Electronic Inspections
RCT ID
AEARCTR-0004458
Initial registration date
July 29, 2019

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
August 02, 2019, 3:39 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Universidad del Pacifico

Other Primary Investigator(s)

PI Affiliation
UC Berkeley
PI Affiliation
International Finance Corporation
PI Affiliation
World Bank

Additional Trial Information

Status
In development
Start date
2019-08-05
End date
2019-12-31
Secondary IDs
Abstract
Business safety inspections (BSIs) are one of the most commonly cited barriers to doing business. Costs imposed by BSIs include uncertainty about the requirements to obtain a certificate (caused by the discretion in the application of the regulation), long waiting times (both for inspections and the associated administrative process) and leakage. In this trial we assess the effects of introducing electronic inspections on outcomes related to inspector productivity. We measure impact on inspector productivity (measured by the average duration of inspections and number of inspections per day), inspection process efficiency (measured as time to process paperwork from application to obtaining the certificate), and standardization and discretion (measured by variation in compliance rates at the item- and section-levels between paper and electronic forms).
External Link(s)

Registration Citation

Citation
Barron, Manuel et al. 2019. "Building Safety Inspections: Paper vs Electronic Inspections." AEA RCT Registry. August 02. https://doi.org/10.1257/rct.4458-1.0
Former Citation
Barron, Manuel et al. 2019. "Building Safety Inspections: Paper vs Electronic Inspections." AEA RCT Registry. August 02. https://www.socialscienceregistry.org/trials/4458/history/51132
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
Business safety inspections in Peru are conducted by municipalities with their own staff. Currently, these inspections are conducted on a paper form containing more than 100 items regarding compliance with regulation of hallways, electronic boards, staircases, evacuation routes, signaling, etc. Inspectors visit a business, walk around the premises, take notes, and at the end of the inspection usually fill the form from memory, assisted by notes taken during the walk-through.

There are three types of inspections. Low or medium risk (conducted by one inspector); high risk (two inspectors of different specialties); and very high risk (three inspectors of different specialties). At the end of inspections with two or more inspectors, inspectors need to compare notes and fill out the items corresponding to the area in which they specialize. This discussion takes time and is complicated by the fact that some items are not clearly defined as pertaining to one or another inspector (like signaling).

The paper form contributes to making the process highly discretionary. Some items in the checklist require to inspect multiple units of the same element, such as electric boxes, and then answering whether all the units comply. This leaves room for discretion, interpretation, and error. For example, an establishment might have 20 electric boxes and only one that is non-compliant. The form itself has no way of showing this rate of non-compliance, let alone the number of electric boxes in the premises. Two inspectors applying the same criteria might therefore reach different compliance outcomes, depending on whether they inspected the one box that was non-compliant. In addition, inspectors might apply different criteria, with some considering that 95% compliance rate for an element should be awarded a “comply” in the form, and others considering that non-compliance of a single unit deserves the element to be marked as “does not comply”. In high and very high risk inspections (two or more inspectors) the paper form can also lead to error. Because multiple inspectors share only one paper form, at least one of the inspectors has to rely on their memory during the walk-through, which means they might pass over certain items. Business owners usually complain about this lack of predictability, and some have to go through up to even five inspections before obtaining a certificate.

This study is embedded within a larger experiment aiming at improving the inspection system. This larger intervention consists of introducing electronic business safety inspections in four municipalities. By using tablet computers with software specially developed for the inspections, inspectors will fill out the forms electronically as they conduct the visit. In inspections that require multiple inspectors, each inspector will have access only to the sections that correspond to their area of specialization, and they will merge their reports seamlessly at the end.

Our intervention will provide a more standard data entry instrument. In addition, it will generate compliance reports automatically, as well as other reports that were previously done by the inspector after the inspection, saving time to the inspector and the firm. Furthermore, electronic inspections will increase accountability, as all inspection data and results will be automatically uploaded to a dashboard that can be accessed by the inspectors’ supervisor.

Our intervention also reduces the problem of inspector discretion in instances of multiple elements. For example, the electronic system includes a space to identify each electric box by an ID, and then conduct the inspection box by box. Furthermore, some items in the paper form pool together hallways and staircases. In the electronic form, the inspector can enter data for each hallway and staircase separately. At the end, the system produces a report with the same format required by law, but the municipality has more disaggregated data, which can allow them to identify compliance problems more accurately.
Intervention Start Date
2019-08-05
Intervention End Date
2019-12-31

Primary Outcomes

Primary Outcomes (end points)
Inspector productivity (measured by the average duration of inspections and number of inspections per week) and inspection process efficiency (measured as time to process paperwork from application to obtaining the certificate).
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Inspection outcomes (number of violations, overall and per section) and the inspectors’ learning curve.
Secondary Outcomes (explanation)
Learning will be measured by comparing treatment effects of the outcome variables at biweekly intervals. For instance, treatment may initially increase duration of the inspection and decrease it towards the end of the trial, as inspectors get familiar with the electronic system.

Experimental Design

Experimental Design
We will randomly sort inspectors into two groups: treatment and control, stratifying by municipality and technical background of the inspectors (electricity, architecture or civil engineering, safety, and “basic”). Stratification by inspector background allows us to assess the impact of electronic inspection on team inspections.

Inspectors in the control group will conduct their inspections on paper (status quo), while inspectors in the treatment group will conduct their inspections using tablets and following the protocol for electronic data collection.

Impact will be estimated comparing treatment and control outcomes for the 6 weeks of the duration of the intervention.
Experimental Design Details
Randomization Method
Randomization done in office by a computer
Randomization Unit
The unit of randomization is the inspector.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
40 inspectors
Sample size: planned number of observations
The unit of observation is the inspection. If each inspector conducts 4 inspections per day, there would be 120 visits on average over a 6-week period. Thus, with 40 inspectors we would have 4,800 observations.
Sample size (or number of clusters) by treatment arms
20 inspectors in treatment, 20 inspectors in control arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
With 20 inspectors per arm, observed in 120 inspections each, correlation between follow-up measurements=0.05; the MDE is 0.22 standard deviations.
IRB

Institutional Review Boards (IRBs)

IRB Name
IRB Approval Date
IRB Approval Number

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials